Correlation

• Correlation- the strength and direction of an association between two or more co-variables. Plotted on a scatter diagram. Can have a positive, negative or no correlation. Correlations and experiments are different as experiments manipulates the IV to measure the DV, assuming cause and effect. In a correlation there isn't any manipulation on one variable, not possible to establish cause and effect, can't assume caffeine causes anxiety for example. 
• -Strengths, it is quantifiable, they assess the strength and direction of a relationship, showing a pattern which can be easily analysed. Correlations are usually used as a starting point of an experiment. They are quick and quite cheap, no need for a controlled environment or manipulation of variables. Data collected by others can also be used. 
• -Limitations, doesn't explain or show cause and effect, we don't know which variable causes the other to change. An untested variable may be causing a relationship between variables, an intervening variable or third variable problem. May be misused or misinterpreted. E.g. a person who works a lot has high anxiety, but they also may drink a lot of caffeine, increasing anxiety levels.